Document of The World Bank

Transcription

1 Public Disclosure Authorized Document of The World Bank FOR OFFICIAL USE ONLY Public Disclosure Authorized Public Disclosure Authorized IMPLEMENTATION COMPLETION REPORT (CPL-41210) ON A LOAN IN THE AMOUNT OF US$30.0 MILLION TO THE STATE OF PIAUÍ FOR A RURAL POVERTY ALLEVIATION PROJECT Report No: /24/2002 Public Disclosure Authorized ESSD Brazil Country Management Unit Latin America and the Caribbean Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.

2 CURRENCY EQUIVALENTS (Exchange Rate Effective December 31, 2001) Currency Unit = Real R$1.00 = US$ 2.32 US$ 1.00 = R$0.43 FISCAL YEAR January 1 to December 31 ABBREVIATIONS AND ACRONYMS APCR CAS CEPISA EMATER FAO/CP FUMAC FUMAC-P GOB IDH IICA INCRA IRR MIS MTR NGO NRDP O&M PAC POA PRONAF R-NRDP RPAP RPRP SEPLAN SOE TU Program of Support to Small Rural Communities Country Assistance Strategy State Electricity Company of Piauí Technical Assistance and Rural Extension Agency World Bank/FAO Cooperative Program Municipal Community Schemes Pilot Municipal Community Funds Federal Government of Brazil Human Development Index Inter-American Institute for Cooperation in Agriculture National Institute of Colonization and Agrarian Reform Internal Rate of Return Management Information System Mid-term Review Non-Governmental Organization Northeast Rural Development Program/Project Operation and Maintenance State Community Schemes Annual Operating Plan National Program to Strengthen Family Agriculture Reformulated Northeast Rural Development Program/Project Rural Poverty Alleviation Project Rural Poverty Reduction Project State Secretariat of Planning Statement of Expenditure Technical Unit Vice President: Country Manager/Director: Sector Manager/Director: Task Team Leader/Task Manager: David de Ferranti Vinod Thomas John Redwood Luis O. Coirolo

3 BRAZIL Rural Poverty Alleviation Project - Piaui CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 4 5. Major Factors Affecting Implementation and Outcome Sustainability Bank and Borrower Performance Lessons Learned Partner Comments Additional Information 25 Annex 1. Key Performance Indicators/Log Frame Matrix 29 Annex 2. Project Costs and Financing 30 Annex 3. Economic Costs and Benefits 32 Annex 4. Bank Inputs 41 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 42 Annex 6. Ratings of Bank and Borrower Performance 43 Annex 7. List of Supporting Documents 44

4 Project ID: P Project Name: Rural Poverty Alleviation - Piaui Team Leader: Luis O. Coirolo TL Unit: LCSER ICR Type: Core ICR Report Date: June 28, Project Data Name: Rural Poverty Alleviation - Piaui Country/Department: BRAZIL Sector/subsector: AY - Other Agriculture L/C/TF Number: CPL Region: Latin America and Caribbean Region KEY DATES Original Revised/Actual PCD: 03/09/1996 Effective: 11/01/ /04/1997 Appraisal: 07/15/1996 MTR: 02/01/ /01/2000 Approval: 12/12/1996 Closing: 06/30/ /31/2001 Borrower/Implementing Agency: Other Partners: STATE OF PIAUI/SEPLAN/Project Technical Unit Municipalities and Community Associations STAFF Current At Appraisal Vice President: David de Ferranti Javed Burki Country Manager: Vinod Thomas Gobind T. Nankani Sector Manager: Mark Cackler Constance Bernard Team Leader at ICR: Luis O. Coirolo Luis O. Coirolo ICR Primary Author: Anna Roumani with project Task Manager (Tulio Barbosa) and Brazil Northeast Team 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) Quality at Entry: Project at Risk at Any Time: No ICR S

5 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The objectives of the Rural Poverty Alleviation Project (RPAP), as stated in the Loan Agreement were to assist the Borrower s efforts to alleviate rural poverty in the Municipalities by: (a) providing basic social and economic infrastructure and employment- and income-generating opportunities; (b) supporting rural community groups in planning and implementing their own subprojects; (c) providing a safety net for the rural poor; and (d) leveraging revenue mobilization at the community and municipal levels. The project was part of the Bank s Program of Targeted Interventions. Important complementary objectives were accurate poverty targeting, i.e., the direct, reliable transfer of funds to the poorest communities, and fostering decentralized decision-making at the state, municipal and local levels. 3.2 Revised Objective: Project Objectives were not revised 3.3 Original Components: The project had three principal components: (a) Community Subprojects (US$35.4 million, 90% of total base cost) supporting small-scale investments of up to US$50,000 each, selected, implemented, operated and maintained by the beneficiaries themselves; (b) Institutional Development (US$2.0 million, 5% of total base cost) providing implementing entities and communities with technical assistance and training to increase their capacity and improve project implementation; (c) Project Administration, Supervision, Monitoring and Evaluation (US$2.0 million, 5% of total base cost) financing project coordination and activities to assess project performance and impact. The Bank financed 75% of total estimated cost, and the State Government, municipalities and beneficiaries the remaining 25%. These components incorporated an appropriate set of actions to achieve project objectives and directly benefit some 810,000 of the poorest rural residents (about 43% of the target population of this state). Table 1 shows project components with cost at appraisal and actual expenditure at Closing, and end-project rating. Project cost was slightly lower than estimated at appraisal due mainly to the effects of repeated devaluation of the Brazilian Real/US Dollar exchange rate starting in Table1. Comparison of Cost at Appraisal and Actual Expenditures Rating Component Total Cost Est. at Appraisal (US$mill.) Expenditures at Closing (US$mill.) S A. Community Subprojects US$35.40 of which: S FUMAC S FUMAC-P S PAC S B. Institutional Development S C. Project Administration, Monitoring, Evaluation and Supervision Physical and Price Contingencies Total: Revised Components: Project components were not revised, but the Loan Agreement was amended as follows: (i) Closing Date extended from June 30, 2001 to December 31, 2001; (ii) Loan funds totaling - 2 -

6 US$2.0 million reallocated from the FUMAC subprogram (1-A) to PAC (1-C) in November 1999; (iii) additional reallocation of US$1.1 million made from FUMAC-P (1-B) to FUMAC (1-A, US$600,000) and Consultant Services (Category 2, US$500,000) in October 2000; and (iv) US$500,000 reallocated from Category C Project Administration for supervision activities, 50% financed by the Bank. 3.5 Quality at Entry: Appraisal of this project pre-dated the QAG. Quality at entry is rated Satisfactory. In conceptual and design terms, the project was a distillation of experiences from several decades of rural development lending in Northeast Brazil, as well as LAC and other regions. The lessons of these efforts have been extensively documented by the Bank and the wider development community, and reflected in diverse ways in many new projects. Participatory, demand-driven mechanisms successfully piloted under the preceding reformulated Northeast Rural Development Program (R-NRDP, ), were the project's core feature. Project objectives were consistent with rural conditions in the State of Piauí, the State s declared strategy for combating rural poverty, and the Bank s most recent CAS for Brazil (1993). Piauí is one of the poorest states in Brazil and certainly the poorest in the Northeast region, as borne out by a range of socio-economic indicators. Skewed access to land, traditional patterns of low productivity subsistence farming and exceptionally high deficits of basic infrastructure and services, largely explains the precarious living conditions of most rural families. At project outset, IPEA s Mapa de Fome showed 60% of families surviving below the poverty line in 90% of the state s municipalities. Further, 97% of rural families had no reliable access to water; 93% had no sanitary facilities; and 60% of the rural population over 15 was illiterate. The State Government s declared strategy for combating rural poverty was driven by exceptional levels of rural-urban migration and focused on employment and income-generation in agriculture, agro-industry and rural electrification, along with other efforts in housing and sanitation for the poorest populations. The Brazil CAS (1993) called inter alia, for proper targeting and delivery of basic services to the poor to reduce poverty and inequity. The State Government and the project Technical Unit demonstrated their commitment to the project model and philosophy through successful measures in the latter stages of the R-NRDP to improve problematic performance in preparation for a strong launch of the RPAP, and demand for project investments from the poor communities was exceptionally strong. As in the case of the R-NRDP, the project approach represented a marked shift from several decades of rural development programs in the Northeast which, while piloting important sector initiatives, had only limited impact on rural poverty. The new methodology, piloted as a small component of the NRDP and established, by reformulation of the NRDP in 1993, as a template for poverty projects with greater impact, cost-effectiveness and sustainability, has a set of guiding principles: (i) place the funds for implementing approved subprojects directly into the hands of beneficiary communities; (ii) stress community participation in the planning, financing, execution, operation and maintenance of investments; (iii) decentralize decision-making and involve local authorities as participants in the process; (iv) maintain transparent decision-making and processes at all levels; and (v) use simple, explicit and monitorable poverty targeting mechanisms

7 Administratively simple and dispensing with intermediary agencies and the bureaucratic/coordination stresses of previous programs, this approach ensures that the vast majority of project resources (about 90%) are used for direct subproject investment, with the balance focused mainly on technical assistance and training for beneficiary communities. Three mechanisms of increasing decentralization are used to screen, approve and implement community requests for financing. Under PAC (State Community Schemes), rural communities submit investment proposals directly to the State Technical Unit (TU), which screens and approves them, releasing funds to the beneficiary associations. In the case of FUMAC (Municipal Community Schemes), decisions concerning investment proposals are delegated by the state to project Municipal Councils, representative bodies with 80% of their members coming from the communities and 20% from municipal authorities and local civil society. Councils discuss and establish priorities, and approve community proposals. Reviewed by the TU for consistency with the Operational Manual, proposals are then funded, with disbursement directly to the association bank account. Councils, together with the TU, monitor and supervise subproject execution. FUMAC-P (Pilot Municipal Community Funds) is a more decentralized variant of FUMAC, introduced as a pilot under the RPAP. Working with an annual budget amount established by the TU on the basis of transparent criteria rural population, poverty level and previous performance Councils submit an Annual Operating Plan (POA) for TU review. Approval releases budget to the Council which then manages its distribution to beneficiary associations and assists with subproject execution. The state TU concentrates on monitoring, technical assistance and supervision. Design of the Piauí project, as with all seven other projects within the RPAP, reflected lessons from the R-NRDP, expanding successfully-piloted features and introducing further innovations and improvements. The FUMAC Councils were mainstreamed and FUMAC-P was introduced, giving greater control over resources to selected, high-performing FUMAC Councils; beneficiary targeting and transparency were systematized with incentives and penalties governing performance; the MIS was improved, with data sent regularly to the Bank via internet, to support targeting, monitoring and evaluation, and project management; the Operational Manual was introduced (part of the project legal documents) with technical and financial guidelines; and technical assistance was funded for mobilization, organization and capacity building. Other features included intensive supervision, standardized subproject design and cost parameters, publicity/information campaigns, cost-sharing requirements for the state government and beneficiaries and best-practice exchanges with other states and between participating entities (Councils, community associations). Experience in Brazil and elsewhere had demonstrated that rural poverty projects were most vulnerable to irregular funding, technical and design weaknesses and inadequate supervision, poor targeting of funds and overly-centralized decision-making. Prudent measures were adopted, and successfully employed, to avoid or reduce such risks. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The outcome of the Project is rated Satisfactory

8 The core project objective was to assist the Borrower's efforts to alleviate poverty in rural municipalities through the cumulative impact of the following: Objective Providing basic social and economic infrastructure, and employment and income-generating opportunities: The project met this objective, financing 1,170 small-scale investments in basic rural infrastructure, productive and social facilities, in 200 municipalities, benefiting 996 community associations (and 935 communities)* and about 480,000 people. Infrastructure was about three-quarters of all investments, with rural electrification subprojects (63%) installing some 2,600 km of line benefiting 31,000 families. Water supply investments (a further 30% of infrastructure) benefited another 18,000 families. Among productive investments, agricultural mechanization (community tractors and associated equipment) represented 67% and benefited about 8,300 families. Septic tanks/sanitation systems were 64% of all social investments, and benefited 9,400 families. Despite forecasts at appraisal that community demand would increase for productive and social investments, in practice communities continued to seek the basics and as evaluation has repeatedly shown, these basics are catalysts to other important activities and outcomes, including employment and income generation, better community health and security, educational and leisure opportunities, commercial activity and increased municipal and state fiscal revenues. It is worth noting that the project Baseline Study (1999) found that, compared to family and per capita income data in the 1991 IBGE Demographic Census for Piauí, which showed well over two thirds of rural families with incomes of less than one minimum salary per month, there had been a sharp alteration in the curve by 1999, when these indices had dropped to about 25% of rural families and some 37% of individuals. In both cases, having subprojects from the R-NRDP and/or the RPAP was found to be an important determining factor in these results.** * Some communities had more than one association/subproject. * Estudo de Perfil de Entrada Relatorio Final, SEPLAN/Inter-American Institute for Cooperation in Agriculture (IICA), September Installed rural electrification is illuminating streets, homes, schools, small shops/processing units at an average cost of about US$580 per family. The many benefits include access to information, means of communication and to household appliances including refrigeration. They also generate employment and income from the operation of small businesses, agro-industries and small-scale irrigation systems, and permit evening school for children and adults and social/leisure activities under safer night-time conditions. The vast majority of these investments were executed by individual firms contracted by the associations.* * Contractors prepare the technical subprojects (facilities and budget) which are sent to the State Electricity Company (CEPISA) for technical review. CEPISA determines if there is energy available for the intended location (i.e. supply lines). If so, CEPISA prepares a Technical Feasibility Study (TFS for which associations do not pay) which is then sent by the association with its subproject proposal, to the TU for approval. Once approved, the association contracts a firm to execute the works. CEPISA subsequently carries out operation and maintenance tasks. Water supply investments, whether from surface or groundwater and using tube/artesian wells or small dams/reservoirs, besides delivering obvious benefits for a drought-prone region are also a critical input for much productive activity, especially irrigated agriculture. They also facilitate - 5 -

9 women s work at home, reduce the incidence of water-borne disease with special impact on children and the elderly, improve household hygiene and reduce the health-care costs for municipalities. Reliable data was not available to calculate actual savings from water supply subprojects. However, detailed analyses done for the states of Bahia, Ceará and Sergipe indicate that the cost savings from reduced time spent on daily water collection, disease reduction and provision of fewer water trucks during drought periods, are almost certainly substantial, given the numbers of beneficiaries. Communal tractors and equipment can significantly expand area cultivated by small farmers, ensure the timely preparation of land to exploit scarce rainfall, improve the quality of soil preparation and facilitate cultivation and post-harvest activities, all of which contribute to better incomes and community well-being (see also 4.2). Objective Supporting rural community groups in planning and implementing their own subprojects: The institutional support structure established under the project to help communities plan and implement investments greatly exceeded appraisal estimates and was an impressive achievement, especially in view of the politically difficult circumstances under which FUMAC expansion took place, for example. By Closing, poor rural communities were represented by some 996 associations and 133 FUMAC Councils (see Annex 1). Further, given the modest size of the Piauí TU and absence of regional offices, and the large distances involved in supervising the project, the TU s ability to cover virtually the entire state and to make at least three supervision/monitoring visits to each subproject during execution, testifies to its commitment to beneficiaries and increased the quality of subproject outcome in each case. Objective Providing a safety net for the rural poor: Along with a range of other State- and Federally-funded social programs, project investments have bolstered the safety net to reduce the effects of governments fiscal tightening in the period. As in other participating states, most RPAP investments in Piauí complemented traditional public investments in rural infrastructure and were frequently executed by the communities themselves or by private firms contracted by them. Unit costs were compared for five types of subprojects (small-scale community dam, concrete bridge, rural electrification, manioc mill and sewing workshop) sampled in six states including Piauí. Comparing secondary cost data, unit costs for more than 50% of the components for these types of subprojects were found to be lower, significantly so in the case of manioc mills and sewing facilities due to labor cost savings. These decentralized, participatory projects became, under the R-NRDP and the RPAP, a uniquely important source of funding for protection of the rural poor in the Northeast. Objective Leveraging revenue mobilization at the community and municipal levels: Successful leveraging of revenue mobilization from communities and municipalities has been a feature of the RPAP. Piauí is unique among its peers in actively encouraging municipalities to share counterpart funding for subprojects in their area, the State believing (justifiably) that - 6 -

10 municipalities willingness to contribute counterpart demonstrates commitment and promotes ownership. The practice has been continued under the RPRP, since it seems to work well. The Bank s initial concern was that some municipalities either without adequate resources or with party political pressures might not be able to pay and their poor communities would not have access to the project or would have to obtain materials and equipment from firms in exchange for counterpart payment, and these firms might overstate subproject size. Bank missions always stressed that the State Government, through the TU, was responsible for counterpart funds in cases where the municipality was unable. What now seems clear is that the vast majority of participating municipalities did contribute, even the poorest, and that this did not prompt the kinds of problems initially anticipated.* * The Mayors reason that for an investment of R$15 they leverage another R$85 comprising Loan funds and beneficiaries 10% contribution. However, in a state with such low municipal IDH counts this may cause discrimination in some cases, including where a municipal budget is heavily dependent on Federal or state transfers which do not arrive in a regular sequence or in one parcel. Other states such as Sergipe and Paraíba also use the municipal contribution strategy but to a much smaller - more ad hoc extent, while a larger richer state such as Bahia does not bother with this practice. Actual financial contributions by municipalities, in addition to the customary logistical support provided to the Councils (equipment, premises, transportation to meetings), the time spent by municipal officials attending Council meetings, and other forms of support including technical assistance, totaled US$5.5 million, or just under 15% of total project cost of US$37.6 million. Beneficiaries contributed, mostly in labor and materials, some US$3.3 million, or 8.8% of total project cost. 4.2 Outputs by components: A. Community Subprojects (Actual: US$33.1 million equivalent, 88.0% of total base cost) Rated Satisfactory This component financed small-scale investments selected, implemented, operated and maintained by the beneficiaries. Investments of up to US$50,000 each were financed through matching grants (public grants matched with community contribution) to legally established associations identifying their priorities and committing to operation and maintenance responsibilities through their labor or user fees. The project was expected to finance 1,770 subprojects over four years, benefiting some 162,000 families organized into 885 associations in 110 municipalities. An estimated 53 FUMAC and 5 FUMAC-P Councils were to be established. These targets were largely indicative due to the demand-driven, flexible nature of the project. In just over four years, the Piauí RPAP: (i) implemented 1,170 subprojects (66% of target) of which 883 were infrastructure (75% of total), 127 productive (11% of total) and 160 social (14% of total investments) benefiting about 96,000 families in 200 municipalities*; (ii) financed at least one subproject for 996 associations (113%); and (iii) established 131 FUMAC and two FUMAC-P Councils (247% and 40% of target, respectively). These results are very satisfactory given disruptive political and administrative events in the project s initial years. The difference between actual and expected subprojects is largely explained by far fewer productive investments in a demand-led process and the predominance of rural electrification (63% of all infrastructure), at higher average cost (but larger numbers of beneficiaries), in the mix.* About 17% of associations were repeat participants, having implemented two or more investments;

11 municipalities received 11 or more subprojects and three received over 20, the latter attributed to old-established (R-NRDP), well-organized and agile FUMAC Councils with a large number of associations and efficient subproject execution. The vast majority received one subproject in an attempt by FUMAC Councils to equitably maximize coverage. * Lower than expected family numbers are due to over-estimation at appraisal, fewer subprojects overall, and rural distances and population density in Piaui. ** TU technicians believe this may continue even under the RPRP as Councils seek full coverage. Access to basic infrastructure and services remains poor in Piauí and the quality of the labor force for successful productive ventures is still relatively low. The Physical Performance Study (1999) conducted by the TU for the Mid-term Review, analyzed a sample of 42 subprojects (rural electrification, community tractors and equipment, water supply and sanitation systems), or about 8% of the total implemented in Piauí at that point (519). A majority of subprojects were adequately sized for community needs; most investments were benefiting association members and non-members, all of whom were paying user fees as appropriate; functionality ranged from excellent to good/regular in 65% of cases, the remainder being at too early a stage of start-up to properly assess function. In addition, a September 2001 Bank supervision mission conducted both field and desk assessments of a diverse sample of subprojects, finding 100% concluded and operational, and that levels of community participation, subproject quality and quality of community supervision of their execution, were fully satisfactory in 90% of cases and good in the other 10%. Piauí s performance in establishing FUMAC Councils greatly exceeded appraisal expectations, as stated above, and its commitment to the methodology is commendable. Councils development was not entirely smooth however, and with hindsight, given the evolving socio-economic and political context of Piauí, this should not surprise. About 50% of subprojects approved in the first two years were PAC, a pattern common across all participating RPAP states. A Bank-prompted technical audit of all PAC subprojects by a new TU administration in 1998 resulted in an agreed program of corrective actions to reduce the large number of non-compliant community associations. PAC was deactivated and FUMAC expansion accelerated due to the state s belief that it delivered superior results. By end-project some two thirds of all investments were FUMAC-based. Rapid establishment of Councils during the expansionary period of 1998/99 however, some without adequate oversight by the TU, created distortions. Internal processes were not adequately participatory; there was early evidence of political influence on members and lack of transparency in approvals of some subprojects, affecting their cost, size and operation. Numbers of subprojects being proposed were limited by Councils weak understanding of the Operational Manual and the Project itself, due to the patchy dissemination of information. Again, an evaluation was recommended by the Bank and effectively carried out by the TU, with positive results. Councils were unable to obtain subproject financing until they were judged fully satisfactory in representational and operational terms. Based on the excellent working relationship between the TU and the Bank Task Manager, problems affecting the Councils performance were continuously monitored and remained on the table, with the TU committed to upgrading its performance albeit with some ongoing philosophical differences with the Bank as to Councils true readiness and capacity to carry out - 8 -

12 their full mandate, especially the idea that they could handle more complex tasks/responsibilities. Bank concerns linger about the autonomy of some Councils, social capital development and an element of residual paternalism in the TU s oversight of Council activities/performance, but impressive progress was made and this is likely to continue under the RPRP. While the project funded a large number of courses and training events for the Councils and associations, in Council/community management, associative skills, operation and maintenance and practical, productivity-related skills, a more systematic delivery of technical assistance and training, specifically in organization, mobilization and consciousness-raising and especially for the Councils, remained an objective. Bank-recommended expansion of partnerships with private and public agencies, NGOs and the Church, was slow to materialize and will benefit from renewed efforts under the RPRP. Other forms of support designed to maximize subproject quality, e.g., the dissemination of revised, updated standardized designs, the application of more stringent standards to productive subprojects, and monitoring the performance of service providers, continued to have issues cited in Bank supervision reports, although here too, there had been major improvement by end-project. Agreed action plans were generally fully implemented, but some more complex issues arising in the latter stages of the project were, by agreement between the Bank and the TU, carried over to the new project. The TU fully understands and acknowledges what needs to be done.* * As an interesting note, a SEPLAN end-project paper (April 2002) briefly discusses the need longer-term, to avoid assistencialismo caused by the quasi permanent presence of NGOs, national and international bodies, the Church and government, working at the local level on behalf of associations and Councils, citing the difference between providing the fish and teaching fishing. Average subproject cost at Closing was about US$27,000 (well below the US$50,000 ceiling) with substantial variations depending on type of investment. The overall average per family benefited was about US$340. (Cost/beneficiary numbers can be a function of families counted as beneficiaries rather than absolute cost of the investment). Average cost differed significantly between PAC and FUMAC infrastructure and social subprojects and less so, for productive. The per capita cost of rural electrification investments in Piaui was found to be about 30% lower than similar subprojects executed by the public sector. Though less common, water supply subprojects showed a similar trend. Average costs are shown below. Average Subproject Cost, Piauí (R$)* Subprogram Infrastructure Productive Social Total FUMAC 26,274 30,195 25,094 26,436 PAC 31,117 29,439 32,758 31,030 Total 27,877 29,783 27,738 28,077 *Average exchange rate for the project period was R$1.66/US$1.00. Targeting was especially effective under the Piauí project. Some 94% of subprojects benefited municipalities with an Index of Human Development (IDH)* below and 62% were in municipalities below Further, at the time of the Mid-term Review, according to the coefficients of the Municipal Participation Fund (FPM), 85% of subprojects financed were for municipalities with a coefficient below 1.6. This data supports the conclusion that project targeting was successful. Further, experience with the Councils shows that the communities - 9 -

13 themselves are best able/informed to target effectively. Women benefited especially from the large number of investments in electrification and water supply. * United Nations composite index measuring economic and social development. B. Institutional Development (Actual: US$3.6 million equivalent, 9.6% of total base cost) Rated Satisfactory This component provided technical assistance and training to the Councils, associations and the TU for capacity-building and improved project execution, and to analyze issues surrounding modernization and reform of the State. The program comprised seminars, workshops and a large number of specific training events in the field, benefiting the TU itself, Councils and associations. The Inter-American Institute for Cooperation in Agriculture (IICA) was a key agent in this effort, supporting the TU under contract to SEPLAN. Project financing (100%) for consultancies, studies, technical assistance and study tours, along with acquisition of computer equipment, had the following impact on the four areas defined as priorities under the TA subcomponent for modernization and reform of the state: (i) Voluntary Separation: some 8,540 state employees were released, reducing the state s salary obligation by R$2.5 million per month; (ii) Control of the folha de pagamento: views on impact differ. According to the TU, the effects were modest, the real impact being ethical. Other information however, shows the state moving from a primary deficit of R$2.0 million in 1997 to a surplus of R$157.0 million in 1999, citing project-funded actions as a major reason; (iii) Privatization: the planned privatization of CEPISA, the State Electricity Company, was started but remains incomplete. Other state holdings, e.g., the State Bank of Piauí (BEPI) were identified and slated for privatization; and (iv) Debt Management: the state s public debt was refinanced by the Federal Government to permit longer-term payback, reducing the monthly service burden. Another important outcome was installation of the Integrated Accounting System to modernize state administrative machinery. Additional State reform activities are included in the new RPRP. C. Project Administration, Monitoring and Evaluation (Actual: US$0.87 million equivalent, 2.3% of total base cost). Rated Satisfactory This component performed well, supporting project coordination and activities to assess/measure project performance and impact; incremental operating costs (excluding salaries) of the TU to enhance supervision and strengthen project operations; the project MIS, evaluation and performance studies and the state s project information campaigns. While expenditures for supervision by the TU were somewhat lower than expected, they financed an intense level of supervision during the subproject implementation cycle; the first to ascertain the legitimacy of community demand, a second during execution and the third at the conclusion of works. 4.3 Net Present Value/Economic rate of return: The SAR did not include an economic or financial rate of return for the project. Instead, an analysis was provided to illustrate the kinds of benefits achieved under the R-NRDP for the most common types of investments. Similar benefits were expected on a larger scale from each of the state projects, including Piauí. Subprojects financed by the RPAP did not differ markedly from

14 the R-NRDP. A similar analysis has therefore been done of the program-wide investments under the RPAP for cost-effectiveness and sustainability, and the financial viability of productive subprojects. Conclusions about fiscal impact are also suggested. It is concluded that the CDD approach to rural infrastructure and service delivery to the rural poorest can work cost-effectively in the Northeast, supported by several basic design features: (i) establishing priorities and selecting investments within an open democratic forum, the Councils; (ii) using standard technical designs and cost parameters, ensuring the use of least-cost models and established patterns of initiating and completing subprojects; and (iii) delegating implementation to the associations, which generates cost-savings relative to similar works executed by public agencies. The contracting procedures in the Operational Manual require direct contracting through competitive processes on all subprojects; associations obtain three price quotations and select the least-cost. Analysis of a random sample of subprojects including ten categories representing about 80% of types financed, found RPAP costs to be about 30-50% cheaper than subprojects of similar quality, implemented by the state. In the six states surveyed, including Piauí, the majority of subprojects were rated technically satisfactory and of good quality. Operation and maintenance has been good, with associations charging appropriate user fees. In a separate, general evaluation in 2000, some 89% of a sample of 8,123 R-NRDP (1995) and RPAP (1997/98) subprojects had remained fully operational. There was no substantial difference in sustainability across infrastructure, productive or social investments. 4.4 Financial rate of return: Internal Economic Rates of Return (IRR) IRRs were estimated for the most common types of productive subprojects financed under the program. These all exceed 30%, an indication of the overall profitability of productive ventures normally financed under the RPAP. Evidence also shows that benefits are largely concentrated in the beneficiary communities. The analyses assumed constant benefits over a ten-year subproject life. Financial Viability of Productive Subprojects Benefit cost ratios are high (greater than 2.0) and investments are financially sustainable, based on analysis of 12 selected, typical productive subprojects. While beneficiaries receive a one-time investment matching grant from the RPAP, it is financially sustainable because cost recovery through user fees is generally adequate to cover O&M and replacement of the original investment long before the end of its useful economic life. All subprojects show satisfactory internal rates of return, some of them being high. Less than six years are needed to recover the investment in all cases and less than 3.5 years in half the cases. IRRs compare favorably with the real cost of borrowing to the Brazilian Government, which at the time of the analysis was about 11-12%. Testing financial robustness of the subprojects, a sensitivity analysis was carried out on the illustrative models which determined that: some subprojects generally those highly dependent on purchased inputs (fish farm, dairy plant, bakery) are more sensitive than others to changing market conditions. In all cases, the worst scenario is one where prices decrease

15 Fiscal Impact Evaluations in the states of Bahia, Ceará and Sergipe indicate that subproject investments make a significant fiscal contribution through cost savings to state and municipal governments and tax revenues. In Piauí, insufficient reliable data was available to make robust calculations but given the large numbers of beneficiaries of water supply and rural electrification subprojects, the fiscal impact was almost certainly substantial. These benefits are, in the case of water subprojects, most commonly-derived from fewer water trucks in time of drought and reduced public health costs from reliable, clean water supply. In the case of rural electrification, the main fiscal impact comes from a large increase in the purchase and use of domestic appliances when electricity became available, increasing the ICMS revenue to municipalities.* Cost savings also derive from the community-driven design of the project; subprojects implemented by communities, directly or contracted, were 30-50% less than prices paid by public authorities for works of similar type and quality. * ICMS is collected by the state and all municipalities receive 25% of collections regardless of where it is collected. 4.5 Institutional development impact: As a participatory, demand-oriented mechanism, the Piauí project, like its peers in other states, involved a range of actors and stakeholders. Evaluations have included a community participation study in six RPAP states, including Piauí, which analyzed social capital formation, and assessed institutional impact of the project on the community associations, Municipal Councils and Technical Units. Also relevant is the tacit recognition of the importance of the RPAP s decentralized model in the (draft) State Economic Memorandum (SEM) for Piauí. The following summarizes findings from both. Community Associations: Under the RPAP, community associations have generally performed well, with members typically meeting their collective responsibilities and benefiting from the efforts of mayors, the state technical units and civil society. In Piauí, associations lacked a history of associativismo; some 57% of the project-based associations are relatively new.* In the initial years weakly organized, lacking knowledge of the project and of FUMAC and with suspect legitimacy in some cases, their function and sustainability challenged the TU. Much has changed. Associations have been encouraged to be family-based and pluralistic. Female-headed associations are slowly increasing and these have the reputation for being well-run; community monitoring committees are mostly female. Subproject benefits have tended, as in other states, to reach members and non-members alike and this has built support for the associations. While broad-based sustainability remains the vision in Piauí, rural communities now commonly view their association in institutional terms as indispensable for demanding/obtaining a range of needed services. Experience shows that technical assistance is critical for institutional success and sustainability of associations at their formative and organizational stages, throughout the investment cycle and longer-term. * Brazilian law requires that communities legally associate to use public funds and several states of the Northeast have, like Piauí, weak associative traditions. The creation of new associations to meet participation criteria is quite logical and while their sustainability can be fosteredin many ways, it cannot be guaranteed

16 Municipal Councils: As fora for democratic discussion, the project Municipal Councils have increased transparency, reduced political interference, improved targeting and built social capital in poor rural communities. Little doubt remains concerning Councils superior effectiveness to PAC at inducing social and cultural change. FUMAC-P Councils performed even better than FUMAC Councils across participating states, generally validating the benefits of deeper decentralization of program delivery. The six-state survey showed that Councils were being integrated into local government planning, steadily overcoming the risk of being short-lived parallel bodies. They are increasingly acknowledged by local populations and officialdom as valid representatives of communities and civil society, and as partners of local administrations in fostering and sustaining municipal development. In Piauí, and as stated in 4.2, the institutional development of FUMAC Councils was problematic, but the sheer numbers of Councils established following the deactivation of PAC was a major achievement in a state with acute poverty, fiscal dependency and a tradition (weakening) of party political influence in the municipalities.* But the institutional effectiveness of Councils in Piauí was diluted by their uniformly small size (range of 9-15 members, with majority 9), an issue whose importance is acknowledged by the TU and which is slated for resolution under the new RPRP. Further, FUMAC-P has evolved slowly. While its annual budget envelope feature has generally fostered even tighter priority-setting, and greater community capacity to identify, plan and implement subprojects in the two FUMAC-P Councils created (Campo Grande and Caxingó), evidence suggests communities were reluctant to assume the required direct management, being satisfied with knowing their indicative budget for planning and decision-making purposes, under the preferred FUMAC. Further efforts are planned under the RPRP to establish and train FUMAC-P Councils. The logistical and financial support of the Mayors is crucial for Councils operations, sustainability and popular legitimacy, and in Piauí this support was extremely strong, including the Mayors contribution of the state s 15% counterpart share in a high proportion of cases, a notable example of fiscal decentralization in practice. Even the poorest municipalities participated in this arrangement, recognizing the opportunity to influence the decision-making process in the Councils. With a relatively small amount of their own resources, they leveraged significant external (State) and local (community) funds. Almost invariably in Piauí, the Council President is an official representing the Mayor, importantly because the public assumes, usually correctly, that such persons have better management skills and access to technical, administrative and other services not necessarily available to it. As with the associations, TA and training to foster members understanding of the Councils wider role beyond Bank-supported investments is judged essential for their institutional growth and sustainability and is being further strengthened under the new project. Technical Unit: The TU administers the project, acts as a conduit for communication between the Bank and State Government, and is the main liaison with the FUMAC and FUMAC-P Councils and in PAC

17 municipalities, with the community associations. Serious institutional issues in the TU during the latter stages of the R-NRDP and administrative and political disruption in the initial phases of the RPAP, were successfully overcome. The unit now has real political, institutional and (relative) budgetary autonomy, as well as strong technical and financial capacity. It has benefited from technical assistance (principally from IICA), is administratively stable, and has adequate vehicles and equipment. Also commendable are its bureaucratic and cost efficiency, ability to rapidly process subproject proposals, professionalism, credibility and effective use of the MIS. Its esprit de corps is remarkable given low salaries and per diems (the latter recently increased modestly), while the quality of its relationships and collaboration with the State Governor, the Mayors, Councils, communities and civil society are noteworthy. Social Capital: Fostering social capital in rural areas and communities has been a striking achievement of the RPAP projects but has varied between states. Several of the more advanced Councils in Piauí became involved in the broader municipal planning process, prioritization and selection of subprojects was conscientious, community ownership strengthened over the course of project implementation, and party-political disputes came to be virtually ignored by the communities by end-project. But in general, social capital development in Piaui remained modest under the RPAP and greater emphasis on this issue is expected under the RPRP. Finally, validation of the RPAP approach from an institutional perspective can also be found in the draft SEM for Piaui (June 2002), as follows: The report urges an increasingly enabling role for the State Government, as opposed to one in which the State Government tries to select the specific economic activities that offer the greatest development potential. This represents a change of philosophy from the past. The report also urges broader modes of consultation and participation in public decision-making and increased transparency through public participation in decisions over policies and investments. With investment resources scarce and the role of the state evolving towards the provider of rights and the setter of ground rules, the rationale for decisions that affect the poor needs to be clear, and the poor must be provided with a voice in such decisions wherever this is technically possible. This entails building up consultative processes in the State (such processes are already present in increasing numbers) with both organized civil society and un-mobilized groups of interested parties..many past development initiatives (normative) dictated priorities and technical solutions and failed through lack of flexibility and dialogue with stakeholders. Government investment programs have since shifted towards more flexible modes (indicative) including wide consultation and participation of beneficiaries. Here, the PCPR* in the Northeast and the program of microbacia ** projects in the South are good examples..the community councils that have been established (under the PCPR) in participating communities may form the basis to solicit further input from these communities into public decisions.there is nothing to stop the articulation of these existing initiatives with new forums for consultation and civil-society participation. PCPR committees, for example, may serve as the basis for local-level consultations on land or water rights. Existing government consultations may also be built-upon. But it will be

18 of central importance to involve beneficiaries in the elaboration of the details of many of the recommendations outlined in this report. *** * Commonly-used, generic name for the Bank-supported rural poverty projects in the Northeast (Programa de Combate à Pobreza Rural) ** Micro-catchment. *** Piauí Managing a Natural Inheritance, State Economic Memorandum, Draft Decision, June 6, Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: The main factor beyond the control of government or the executing agency was the severe El Niño-induced drought of , which affected a broad swathe of the Northeast. The result was generalized productivity losses, agro-processing halted for lack of materials and tractors idle due to overall planting and harvesting conditions. However, as agreed between the Bank and TU in the months prior to the drought, a certain priority was given to the approval of water supply subprojects, with positive effects on livestock survival, alleviation of human suffering and on the fiscal savings to state and municipal governments, of providing fewer water trucks.* * Not quantified but evidence from other states and informal evidence in Piauí strongly suggests positive fiscal impact. 5.2 Factors generally subject to government control: Two issues are relevant. First, the RPAP project was drawn in 1999 into a volatile political dispute stemming from the gubernatorial elections of Complaints about the project by a Federal Deputy were investigated by the Bank project Task Manager and found to relate exclusively to the R-NRDP, to be largely politically motivated, and to have already been detected during routine supervision of that project. At that time (1996), Bank missions demanded corrective action, agreed with the State on an Emergency Work Plan and alerted the State Governor to poor project administration. The project executive director was dismissed as a direct result of Bank complaints and offending associations were prosecuted by the State Attorney General s Office. The other issue involved rural electrification subprojects, for which demand was exceptionally high, and which ran into serious difficulties involving CEPISA (Companhia Energetica do Piauí), the state electricity company, under privatization at the time (and not fully completed yet). While CEPISA defined the problem as one of the RPAP constructing unsafe and technically inadequate distribution lines, government asserted that CEPISA was discriminating against poor clients, including RPAP beneficiaries. The real issue was supply and CEPISA s capacity to fulfill its mandate under Government s Luz do Campo Program. CEPISA was unable to hook up small, remote communities far distant from its distribution lines, or to extend those lines, which left many communities with unfinished subprojects and dashed hopes. All except a few cases (still in process) were fully resolved, due to persistence of the TU and the Bank project Task Manager. 5.3 Factors generally subject to implementing agency control: A number of factors within the responsibility/control of the SEPLAN/Project Technical Unit were detected by Bank supervision and Mid-term-Review missions. These are outlined in 7.2. Most